Artificial intelligence models from Google, OpenAI and Anthropic resorted to nuclear escalation in 95 per cent of simulations ...

According to Payne, the models escalated to the point of tactical nuclear war in 95 percent of scenarios, noting that nuclear threats to each other were far more likely to escalate than de-escalate ...

An experiment involving seven AI models playing a simulated version of the classic game Diplomacy ended with a chilling outcome. OpenAI’s ChatGPT 3.0 emerged victorious—but not by playing fair.